<<

ThePreparationof t-Butyl , , and StyreneBlockCopolymersbyAtomTransferRadical :PrecursorstoAmphiphilicandHydrophilic BlockCopolymersandConversiontoComplex NanostructuredMaterials

QINGGAOMA,KARENL.WOOLEY DepartmentofChemistry,WashingtonUniversity,OneBrookingsDrive,St.Louis,Missouri63130-4899

Received4August2000;accepted8September2000

ABSTRACT: Atomtransferradicalpolymerizationconditionswithcopper(I)bromide/pen- tamethyldiethylenetriamine(CuBr/PMDETA)asthecatalystsystemwereemployedfor thepolymerizationoftert-butylacrylate,methylacrylate,andstyrenetogeneratewell- definedhomopolymers,diblockcopolymers,andtriblockcopolymers.Temperaturestudies indicatedthatthepolymerizationsoccurredsmoothlyinbulkat50°C.Thekineticsof tert-butylacrylatepolymerizationundertheseconditionsarereported.Well-definedpoly- (tert-butylacrylate)(PtBA;polydispersityindexϭ1.14)andpoly(methylacrylate)(PMA; polydispersityindexϭ1.03)homopolymersweresynthesizedandthenusedasmacroini- tiatorsforthepreparationofPtBA-b-PMAandPMA-b-PtBAdiblockcopolymersinbulkat 50°Corintolueneat60or90°C.Intoluene,theamountofCuBr/PMDETArelativetothe macroinitiatorwasimportant;atleast1equivofCuBr/PMDETAwasrequiredforcomplete initiation.Typicalblocklengthswerecomposedof100–150repeatunitspersegment.A triblockcopolymer,composedofPtBA-b-PMA-b-PS(PSϭpolystyrene),wasalsosynthe- sizedwithawell-definedcompositionandanarrowmolecularweightdispersity.The tert-butylestersofPtBA-b-PMAandPtBA-b-PMA-b-PSwereselectivelycleavedtoform theamphiphilicblockcopolymersPAA-b-PMA[PAAϭpoly(acrylicacid)]andPAA-b-PMA- b-PS,respectively,viareactionwithanhydroustrifluoroaceticacidindichloromethaneat roomtemperaturefor3h.Characterizationdataarereportedfromanalysesbygelperme- ationchromatography;infrared, 1HNMR,and 13CNMRspectroscopies;differentialscan- ningcalorimetry;andmatrix-assisted,laserdesorption/ionizationtime-of-flightmassspec-

trometry.TheassemblyoftheamphiphilictriblockcopolymerPAA90-b-PMA80-b-PS98 withinanaqueoussolution,followedbyconversionintostablecomplexnanostructuresvia crosslinkingreactionsbetweenthehydrophilicPAAchainscomprisingtheperipheral layers,producedmixturesofsphericalandcylindricaltopologies.Thevisualizationandsize determinationoftheresultingnanostructureswereperformedbyatomicforcemicroscopy, whichrevealedveryinterestingsegregationphenomena.©2000JohnWiley&Sons,Inc.J PolymSciA:PolymChem38:4805–4820,2000 Keywords:atomtransferradicalpolymerization(ATRP);pentamethyldiethylenetri- (PMDETA);tert-butylacrylate;methylacrylate;styrene;amphiphilicblock copolymer;triblockcopolymer;shellcrosslinkednanostructures

INTRODUCTION ableinterestbecauseoftheirabilitytoself-as- sembleinsolutionorbulkstates.Theself-assem- Blockcopolymerscomposedofsegmentswithdif- blycarriedoutinsolutionallowsforthe ferentsolubilitypropertieshavebeenofconsider- preparationofindividualnanoscopicmaterials generallypossessingacore–shellmorphologyand Correspondenceto:K.L.Wooley(E-mail:klwooley@artsci. beingspherically,cylindrically,orvesicularly wustl.edu) shaped.1,2 Theassemblyofamphiphilicdiblockor JournalofPolymerScience:PartA:PolymerChemistry,Vol.38,4805–4820(2000) ©2000JohnWiley&Sons,Inc. triblockcopolymerswithinanaqueoussolution

4805 4806 MA AND WOOLEY has been followed by crosslinking reactions selec- radical polymerization,24–26 and reversible addi- tively throughout the core3 or shell4–10 domains tion fragmentation transfer polymerization.27–30 to afford stable nanoscale particles. These com- ATRP has been successfully employed for the po- plex nanostructures have shown promise for uti- lymerization of a variety of acrylate monomers, lization in environmental, biomedical, and mate- such as methyl acrylate (MA),31,32 n-butyl acry- rials applications, including the sequestration of late,23 and 2-hydroxyethyl acrylate.33 Under cop- pollutants, drug delivery, gene therapy, coatings, per-mediated ATRP conditions, polymerization is and composites.11–14 initiated by an alkyl halide,34–38 and the revers- The crucial component in each of the supramo- ible homolytic bond cleavage of the alkyl halide lecular assemblies and their stabilized, crosslinked group is facilitated by the redox properties of the counterparts is the nature of the block copoly- copper species as a complex with bidentate or mers, the molecular weight and composition of multidentate N-containing ligands. Among these which dictate the kinetics and thermodynamics of ligands, 4,4Ј-di-n-heptyl-2, 2Ј-bipyridine and 4,4Ј- their organization. We focused on amphiphilic di-(5-nonyl)-2,2Ј-bipyridine have been used exten- diblock copolymers, in which one of the blocks is sively.20,23,34–38 Because these are not commer- poly() (PAA). This allows for the prep- cially available and are relatively expensive to aration of shell-crosslinked (SCK) nanoparticles synthesize, the preparation of large amounts of in an aqueous solution with amidation of the well-defined block copolymers from these ligands acrylic acid groups as the means for crosslinking is hindered. Polymerization in the presence of of the micelle peripheral layer; it also multidentate alkyl amino ligands, such as penta- provides materials of the appropriate morphology methyldiethylenetriamine (PMDETA), has been to be utilized in water.15 The core compositions found to proceed at faster rates and lower tem- and properties in previous examples of SCK nano- peratures.31 Moreover, PMDETA is commercially particles have included mainly hydrophobic chain available and much less expensive than the bi- segments, such as glassy polystyrene (PS),9,15 pyridine derivatives and, thus, deserves further fluidlike poly(isoprene),16 and crystalline poly(⑀- study. Recently, Matyjaszewski et al.39,40 re- caprolactone).17 The incorporation of N-(morpho- ported the polymerization of tert- lino)ethyl methacrylate as the core material gen- (tBA) with the CuBr/PMDETA system. This erated quite interesting structures with tempera- chemistry has also allowed for the preparation of ture-variable core hydrophilicity.5 This prompted diblock copolymers of PS and PAA.41 Because the us to develop an SCK structure based entirely on catalyst is heterogeneous in bulk tBA monomer, hydrophilic PAA, in which the establishment of various polymerization conditions were at- an acrylamido-crosslinked network throughout tempted to determine the optimal conditions for the shell layer would maintain the three-dimen- the polymerization. In particular, the addition of sional architecture. To accomplish the assembly p-dimethoxybenzene, acetone, or N,N-dimethyl- process, however, hydrophobic precursors are formamide as a solvent, together with a 5% excess needed for core nucleation within an aqueous en- of CuBr2 relative to CuBr to moderate the reac- vironment. Therefore, a block copolymer of acrylic tion rate, was found to produce the best results. acid segments for shell formation and crosslink- Reduced temperatures (55 °C or room tempera- ing chemistry must be linked covalently to an ture) in bulk were found to result in slow poly- acrylate segment for core nucleation and merization rates and broad molecular weight dis- organization of the nanoassembly. persities. Although anionic polymerization is an excel- In this article, we report the preparation of lent method for the preparation of well-defined diblock and triblock copolymers of tBA, MA, and block copolymers, it is technically challenging and styrene by ATRP. These block copolymers have not compatible with electrophilic or acidic func- the potential for segment-selective removal of the tional groups. Free-radical polymerization can be ester protecting group because of the reactivity used more broadly with monomers containing differences associated with tert-butyl and methyl functionalities; however, it was not amenable to ester groups and, therefore, serve as versatile the preparation of well-defined , espe- polymer precursors for the preparation of am- cially block copolymers, until the recent break- phiphilic and entirely hydrophilic nanostructured throughs in living free-radical polymerization materials. The ATRP-based of chemistries, including transition-metal-mediated tBA and MA were studied, and it was found that radical polymerization,18,19 atom transfer radical the polymerizations could be accomplished on a polymerization (ATRP),20–23 nitroxide-mediated large scale in bulk, without the addition of sol- PtBA-b-PMA-b-PS BY ATRP 4807

vent. By lowering the polymerization tempera- Glass-transition temperatures (Tg) were mea- ture to 50 °C and carefully controlling the poly- sured by differential scanning calorimetry on a merization procedure to minimize the oxidation of PerkinElmer DSC-4 differential scanning calo-

Cu(I) to Cu(II), we obtained well-behaved poly- rimeter. Heating rates were 10 °C/min, and the Tg merizations occurring at reasonable rates to yield was taken as the midpoint of the inflection tan- well-defined polymers of narrow molecular weight gent after the third or subsequent heating scan. dispersities. The initially prepared homopolymers The PerkinElmer instrument was upgraded with then served as macroinitiators for the growth of an Instrument Specialists, Inc. (Antioch, IL) tem- diblock and triblock copolymers from combina- perature program interface-PE, and data were tions of tBA, MA, and styrene monomers. Selec- acquired and analyzed with TA-PC software (ver- tive cleavage of the tert-butyl ester groups pro- sion 2.11; Instrument Specialists). duced acrylic acid chain segments and converted Tapping-mode atomic force microscopy (AFM) the hydrophobic materials into amphiphilic observations were carried out in air with a Nano- diblock or triblock copolymers. The amphiphilic scope III BioScope system (Digital Instruments, triblock copolymers proved to be of particular in- Santa Barbara, CA) operated under ambient con- terest, in that they underwent self-assembly in ditions with standard silicon tips (type, OTESPA- water to form cylindrically shaped nanoassem- 70; L, 160 ␮m; normal spring constant ,50 N/m; blies, in addition to the common spherical nano- resonance frequency, 246–282 kHz). We prepared spheres. These assemblies were transformed into the samples for AFM analysis by depositing a stable and robust nanostructures by intra-assem- 1-␮L drop of the solution (5–500 ␮g/mL) onto mica bly crosslinking within the hydrophilic shell lay- and allowing it to dry freely in air. ers through amidation of the acrylic acid groups The MALDI-TOF (matrix-assisted laser de- after the reaction with diamino crosslinking sorption/ionization time-of-flight) experiments agents. These nanostructures remained sus- were carried out on a Voyager DE-RP (Perseptive pended in an aqueous solution indefinitely and Biosystems, Framingham, MA). A mixture of organized into unusual ordered arrays after 2,4,6-trihydroxyacetophenone, 2,3,4-trihydroxy- transfer onto a substrate by deposition and drying acetophenone, and ammonium citrate (Sigma, St. of the aqueous solutions. Louis, MO) at a 2/1/2 molar ratio, with a 2,4,6- trihydroxyacetophenone concentration of 0.05 M, was used as the matrix. Samples were dissolved ␮ ␮ EXPERIMENTAL in THF at a concentration of 10 pmol/ L. A 0.5- L sample aliquot and 0.5 ␮L of the matrix solution were mixed on a stainless steel plate. Both nega- Measurements tive and positive ions were desorbed with an N2 IR spectra were obtained on a PerkinElmer Spec- laser (337 nm) and accelerated with a potential of trum BX Fourier transform infrared as diffuse 25 kV. The data were collected with a Tektronics reflectance. 1H NMR spectra were recorded as TDS 520A digitizing oscilloscope (Tektronics, solutions on either a Varian Unity 300-MHz spec- Beaverton, OR), transferred to the computer, and trometer or a Varian Gemini 300-MHz spectrom- processed with Perseptive Grams/386 (version eter with the solvent proton signal as a standard. 3.04; Perseptive Biosystems) software. 13C NMR spectra were recorded at 75.4 MHz as solutions on either a Varian Unity 300 spectrom- Materials eter or on a Varian Gemini 300 spectrometer with the solvent signal as a standard. Nitrogen (99.99%) was used for the polymeriza- Gel permeation chromatography (GPC) was tion and storage of materials. Argon (99.99%) was conducted on a Hewlett–Packard series 1050 used during the transfer of reagents inside an air High Performance Liquid Chromatograph (HPLC) bag. THF (Aldrich; 99%) and toluene were dis- with a Hewlett–Packard 1047A refractive index tilled from calcium hydride. MA (Aldrich; 99%), detector: data analysis was done with Viscotek tBA (Aldrich; 98%), and styrene (Aldrich; 99%) (Houston, TX) Trisec GPC Software (version were distilled from calcium hydride and stored ␮ Ϫ 2.70). PS was used as a standard. Two 5- m Poly- under N2 at 20 °C. Copper(I) bromide (CuBr; mer Laboratories PL gel columns (300 ϫ 7.7 mm) Aldrich; 99.999%), copper(I) chloride (Aldrich; connected in series in order of increasing pore size 99.999%), ethyl 2-bromopropionate (2-EBP; Al- (500 Å, mixed-bed E) were used with tetrahydro- drich; 99%), (1-bromoethyl)benzene (Aldrich; 97%),

furan (THF) distilled from CaH2 as a solvent. ethyl 2-bromoisobutyrate (2-EBiB; Aldrich; 98%), 4808 MA AND WOOLEY

and N,N,NЈ,NЉ,NЉ-pentamethyldiethylenetriamine g/mol, number average degree of polymerization (PMDETA; Aldrich; 99%) were used as received. ϭ 62 by 1H NMR): C, 64.94%; H, 9.35%; Br,

1.00%. Calcd. for C630H1080BrO180 (11,600 g/mol, DP ϭ 90 by GPC): C, 65.15%; H, 9.37%; Br, General Procedure for the Polymerization of tBA n 0.69%. Found: C, 64.87%; H, 9.35%; Br, 0.81%. in Bulk A 250-mL Schlenk flask fitted with a stir bar General Procedure for the Polymerization of MA (oven-dried at 110 °C for 48 h and flame-dried in Bulk under vacuum immediately prior to use) and cov- ered with a rubber septum was charged with A 250-mL Schlenk flask fitted with a stir bar CuBr (981 mg, 6.84 mmol) under argon in an air (oven-dried at 110 °C for 48 h and flame-dried bag. The Schlenk flask was connected to a con- under vacuum immediately prior to use) and cov- denser and a double manifold. The flask was evac- ered with a rubber septum was charged with

uated (0.1 mmHg) for 15 min, backfilled with N2, CuBr (1.33 g, 9.25 mmol) under argon in an air and then immersed in liquid N2. tBA (100 mL, bag. The Schlenk flask was connected to a con- 87.5 g, 684 mmol) was added via a gas-tight sy- denser and a double manifold. The flask was evac-

ringe, followed by PMDETA (1.43 mL, 1.19 g, 6.84 uated (0.1 mmHg) for 15 min, backfilled with N2, mmol). After three freeze–pump–thaw cycles, the and then immersed in liquid N2. MA (125 mL, Schlenk flask was then immersed again in liquid 120 g, 1390 mmol) was added via a gas-tight ␮ N2. 2-EBP (900 L, 1.24 g, 6.84 mmol) was added syringe, followed by PMDETA (2.00 mL, 1.60 g, to the flask via a gas-tight syringe. Three freeze– 9.25 mmol). After three freeze–pump–thaw cy- pump–thaw cycles were performed, and the mix- cles, the Schlenk flask was then immersed again

ture was allowed to stir at room temperature for in liquid N2. 2-EBP (1.20 mL, 1.68 g, 9.25 mmol) 10 min to ensure that the mixture became homo- was added to the flask via a gas-tight syringe. geneous. The flask was then placed into a 50 °C oil Three freeze–pump–thaw cycles were performed,

bath to allow for polymerization under N2. Peri- and then the mixture was allowed to stir at room odically, 0.2-mL aliquots were removed via a gas- temperature for 10 min to ensure that the mix- tight syringe for GPC analysis. After completion ture became homogeneous. The flask was then of the reaction, the flask was immersed in liquid placed into a 50 °C oil bath to allow for polymer-

N2 to quench the polymerization. THF (100 mL) ization under N2. Periodically, 0.2-mL aliquots was then added to the flask, and a vortex tech- were removed via a gas-tight syringe for GPC

nique was used to dissolve the polymer. The mix- analysis. The flask was immersed in liquid N2 to ture was then filtered through a 150-mL fritted quench the polymerization. THF (50 mL) was

Buchner funnel containing an Al2O3/celite plug to then added to the flask, and the mixture was remove the copper. The resulting colorless poly- filtered through a 150-mL fritted Buchner funnel

mer solution was concentrated and twice precip- containing an Al2O3/celite plug to remove the cop- itated into 70% /water (2 L) at 4 °C. The per. The resulting colorless polymer solution was white polymer was collected by vacuum filtration concentrated and precipitated twice into 70% and dried under vacuum for 3 days at room tem- methanol/water (2 L) at 4 °C, and the superna- perature to give poly(tert-butyl acrylate) (PtBA). tant was decanted. The precipitate was dried un- NMR ϭ GPC Yield: 70 g (80%). Mn 8000 g/mol, Mn der vacuum at room temperature for 3 days to give ϭ ϭ ϭ 11,600 g/mol, Mw/Mn 1.17. Tg 34 °C. IR: poly(methyl acrylate) (PMA) as a colorless glass. 755, 844, 1157, 1256, 1366, 1443, 1723, 2975, Yield: 65 g (54%). MALDI-TOF (3-indoleacrylic Ϫ1 1 ␦ ϭ 3434 cm . H NMR (CDCl3, ): 1.05 (d, CH3CH acid doped with AgOAc as a matrix): Mn 8598 ϭ NMR ϭ end group), 1.22 (t, CH3CH2O end group), 1.20– g/mol, Mw/Mn 1.030. Mn 8800 g/mol, GPC ϭ ϭ ϭ 1.50 (br, (CH3)3C), 1.24–1.68 (br, meso and Mn 9110 g/mol, Mw/Mn 1.04. Tg 11 °C. racemo CH2 of the polymer backbone), 1.74–1.94 IR: 756, 826, 967, 1069, 1256, 1443, 1728, 2024, Ϫ1 1 ␦ (br, meso CH2 of the polymer backbone), 2.15– 2960, 3440, 3643 cm . H NMR (CDCl3, ): 1.05 2.35 (br, CH of the polymer backbone), 4.05 (m, (d, CH3CH end group), 1.22 (t, CH3CH2O end CH3CH2O and CHBr end groups overlapping) group), 1.28–1.62, 1.76–2.0 (br, meso CH2 of the 13 ␦ ppm. C NMR (CDCl3, ): 27.9–28.0 [(CH3)3C], polymer backbone), 1.62–1.76 (br, racemo CH2 of 35.7–37.2 (␤ carbon of the polymer backbone), the polymer backbone), 2.15–2.38 (br, CH of the ␣ 41.7–42.2 ( carbon of the polymer backbone), polymer backbone), 3.62 (s, OCH3), 4.08 (q, 80.2 [(CH3)3C], 173.6–173.8 (carbonyl C) ppm. OCH2CH3 end group), 4.15 (t, CHBr end group) 13 ␦ ␤ ELEM.ANAL. Calcd. for C434H744BrO124 (8000 ppm. C NMR (CDCl3, ): 35.7–37.2 ( carbon of PtBA-b-PMA-b-PS BY ATRP 4809

␣ the polymer backbone), 41.8–42.2 ( carbon of the 1.74 (br, CH2 of the polymer backbone), 1.20–1.50 polymer backbone), 51.6 (OCH3), 174.5 (carbonyl [br, (CH3)3C], 1.76–1.94 (br, CH2 of the polymer C) ppm. ELEM.ANAL. Calcd. for C404H606BrO202 backbone), 2.15–2.35 (br, CH of the polymer back- ϭ 1 13 (8800 g/mol, DPn 102 by H NMR): C, 55.30%; bone), 3.50–3.65 (br, OCH3) ppm. CNMR ␦ ␤ H, 6.96%; Br, 0.92%. Calcd. for C412H618BrO206 (CDCl3, ): 27.9–28.0 [(CH3)3C], 34.8–37.2 ( car- ϭ ␣ (9110 g/mol, DPn 105 by GPC): C, 55.32%; H, bon of the polymer backbone), 41.1–42.2 ( car- 6.99%; Br, 0.88%. Found: C, 55.50%; H, 6.98%; Br, bon of the polymer backbone), 51.6 (OCH3), 80.2 0.71%. [(CH3)3C], 174.5 (carbonyl C from PMA block), 174.6 (carbonyl C from PtBA block) ppm. General Procedure for the Preparation of PtBA-b- PMA in Bulk General Procedure for the Preparation of PMA-b- PtBA in Toluene A 100-mL Schlenk flask fitted with a stir bar GPC ϭ (oven-dried at 110 °C for 48 h and flame-dried PMA macroinitiator (Mn 9110 g/mol, Mw/Mn ϭ under vacuum immediately prior to use) and cov- 1.04) was frozen in liquid N2 to facilitate the ered with a rubber septum was charged with transfer of a portion (11.0 g, 1.22 mmol) into a ϭ PtBA as a macroinitiator (Mn 11,600, Mw/Mn 100-mL Schlenk flask fitted with a stir bar (oven- ϭ 1.17; 10 g, 0.86 mmol). The flask was evacuated dried at 110 °C for 48 h and flame-dried under (0.1 mmHg) for 15 min and then backfilled with vacuum immediately prior to use) and covered

N2. CuBr (372 mg, 2.59 mmol, 3.0 equiv to the with a rubber septum. The flask was evacuated macroinitiator) was added under argon in an air (0.1 mmHg) for1htoremovethecondensed mois- bag. The Schlenk flask was connected to a con- ture. CuBr (263 mg, 1.85 mmol, 1.52 equiv) was denser and a double manifold. The flask was evac- added under argon in an air bag. The flask was

uated (0.1 mmHg) for another 15 min, backfilled immersed in liquid N2, and toluene (20 mL) and with N2, and then immersed in liquid N2.MA tBA (20.0 mL 17.5 g, 137 mmol) were added via a (40.0 mL, 38.2 g, 444 mmol, 515 equiv to the gas-tight syringe. After three freeze–pump–thaw macroinitiator) was added via a gas-tight syringe. cycles, the mixture was allowed to stir at room After three freeze–pump–thaw cycles, the mix- temperature for 1 h to ensure that the mixture ture was allowed to stir at room temperature for became homogeneous. The flask was then im- ␮ 20 min and then was immersed again in liquid mersed into liquid N2, and PMDETA (400 L, 318 ␮ N2. PMDETA (541 L, 450 mg, 2.59 mmol) was mg, 1.83 mmol, 1.51 equiv) was added via a gas- added via a gas-tight syringe, followed by three tight syringe. After three freeze–pump–thaw cy- freeze–pump–thaw cycles, and the mixture was cles, the flask was placed into a 90 °C oil bath to then allowed to stir at room temperature for 10 allow the polymerization to proceed. After com- min to ensure that the mixture became homoge- pletion of the reaction, the flask was immersed in

neous. The flask was then placed into a 50 °C oil liquid N2 to quench the polymerization. THF (40 bath to allow for polymerization under N2. Peri- mL) was added, and the solution was filtered odically, 0.1-mL aliquots were removed via a gas- through a 150-mL fritted Buchner funnel contain-

tight syringe for GPC analysis. The flask was ing an Al2O3/celite plug to remove the copper. The immersed in liquid N2 to quench the polymeriza- colorless filtrate was concentrated under reduced tion. THF (50 mL) was then added to the flask, pressure and precipitated twice into 70% metha- and the mixture was filtered through a 150-mL nol/water (1 L) at 4 °C. The white polymer was

fritted Buchner funnel containing an Al2O3/celite collected by vacuum filtration and dried under vac- plug to remove the copper. The resulting colorless uum for 3 days to give PMA-b-PtBA as a white solid. NMR ϭ solution was then concentrated and precipitated Yield: 14 g (49%). Mn 12,640 g/mol, GPC ϭ ϭ twice in 70% methanol/water (1 L) at 4 °C. The Mn 18,700 g/mol, Mw/Mn 1.17. (Tg)PMA ϭ ϭ resulting polymer was collected by vacuum filtra- 11 °C, (Tg)PtBA 44 °C. IR: 752, 845, 1187, tion and dried under vacuum at room temperature 1451, 1992, 2966, 3440, 3643 cmϪ1. 1HNMR ␦ for 3 days to give PtBA-b-PMA diblock polymer. (CDCl3, ): 1.24–1.74 (br, CH2 of the polymer Yield: 15 g (31%). MALDI-TOF (3-indoleacrylic backbone), 1.25–1.50 [br, (CH3)3C], 1.76–1.94 (br, ϭ acid doped with AgOAc as a matrix): Mn 14,875 CH2 of the polymer backbone), 2.15–2.35 (br, CH ϭ NMR ϭ g/mol, Mw/Mn 1.09. Mn 13,000 g/mol, of the polymer backbone), 3.50–3.62 (br, OCH3) GPC ϭ ϭ ϭ 13 ␦ Mn 18,500, Mw/Mn 1.13. (Tg)PMA 14 °C, ppm. C NMR (CDCl3, ): 27.9–28.0 [(CH3)3C], ϭ ␤ (Tg)PtBA 44 °C. IR: 752, 845, 1187, 1451, 1992, 34.8–37.2 ( carbon of the polymer backbone), Ϫ1 1 ␦ ␣ 2966, 3440, 3643 cm . H NMR (CDCl3, ): 1.24– 41.0–42.2 ( carbon of the polymer backbone), 4810 MA AND WOOLEY

ϭ 51.6 (OCH3), 80.2 [(CH3)3C], 173.6–173.8 (car- PMA [Mn 18,500, polydispersity index (PDI) bonyl C from the PtBA block), 174.5 (carbonyl C ϭ 1.13; 1.0 g, 0.054 mmol] and CuBr (30 mg, 0.21 from the PMA block) ppm. ELEM.ANAL. Calcd. for mmol) under argon in an air bag. The Schlenk ϭ C612H996BrO228[12,640 g/mol, (DPn)PMA 102, flask was connected to a condenser and a double ϭ 1 (DPn)PtBA 30 by H NMR]: C, 61.20%; H, 8.30%. manifold. The flask was evacuated (0.1 mmHg) Calcd. for C973H1602BrO344 [18,700 g/mol, (DPn)PMA for 15 min, backfilled with N2, and then immersed ϭ ϭ 105, (DPn)PtBA 75 by GPC]: C, 62.43%; H, in liquid N2. Styrene (10 mL, 9.1 g, 87 mmol) was 8.57%. Found: C, 62.26%; H, 8.70%. added via a gas-tight syringe. After three freeze– pump–thaw cycles, the Schlenk flask was then immersed again in liquid N . PMDETA (40 ␮L, 36 General Procedure for the Conversion of PtBA-b- 2 mg, 0.21 mmol) was added via a gas-tight syringe. PMA to PAA-b-PMA Three freeze–pump–thaw cycles were performed, A clean, 100-mL, round-bottom flask fitted with a and the mixture was allowed to stir at room tem- GPC stir bar was charged with PtBA-b-PMA (Mn perature for 10 min to ensure that the mixture ϭ ϭ 18,500, Mw/Mn 1.11; 3.0 g, 12.5 mmol of became homogeneous. The flask was then placed tert-butyl ester), followed by dichloromethane (30 into a 50 °C oil bath to allow for polymerization mL). The mixture was allowed to stir for 10 min to under N2. Periodically, 0.1-mL aliquots were re- dissolve the polymer. Trifluoroacetic acid (TFA; moved via a gas-tight syringe for GPC analysis. 5.0 mL, 7.4 g, 65 mmol, 5.0 equiv to the tert-butyl After completion of the reaction, the flask was ester) was then added. After the mixture was immersed in liquid N2 to quench the polymeriza- allowed to stir at room temperature for 12 h, the tion. THF (40 mL) was then added to the flask, dichloromethane and excess TFA were removed and the mixture was filtered through a 50-mL at room temperature with dry air gently flowing fritted Buchner funnel containing an Al2O3/celite through the flask overnight. The resulting glassy, plug to remove the copper. The resulting colorless light-brown polymer solid was vacuum-dried for 2 polymer solution was concentrated and precipi- days to give PAA-b-PMA (2.5-g yield). For further tated twice into 80% methanol/water (1 L) at 4 °C. purification, the polymer was dissolved in THF The precipitate was collected by vacuum filtration (20 mL), transferred into dialysis tubing (molec- and dried under vacuum for 3 days to give PtBA- ular weight cut-off ϳ 3500), and dialyzed against b-PMA-b-PS as a white solid. NMR ϭ deionized water for 3 days. Lyophilization gave Yield: 1.2 g (12%). Mn 20,000 g/mol, GPC ϭ ϭ PAA-b-PMA as a white powder. Mn 28,700 g/mol, Mw/Mn 1.16. IR: 473, ϭ Yield: 2.0 g (100%). (Tg)PMA 11 °C, (Tg)PAA 543, 700.3, 757, 845, 1164, 1256, 1370, 1448, ϭ 114 °C. IR: 602, 821, 1053, 1179, 1273, 1447, 1600, 1738, 1946, 2929, 3441, 4049 cmϪ1. 1H Ϫ1 1 ␦ 1663, 1752, 2400–3513 cm . H NMR (CD2Cl2, NMR (CDCl3, ): 1.20–2.00 (br, polymer back- ␦ ): 1.12–1.87 (br, CH2 of the polymer backbone), bone), 1.25–1.50br, [(CH3)3C], 2.00–2.42 (br, poly- 2.15–2.45 (br, CH of the polymer backbone), mer backbone), 3.50–3.64 (br, OCH3), 6.26–6.84 13 3.50–3.62 (br, OCH3) ppm. C NMR (CD3OD/ (br, meta-H from the aromatic ring), 6.84–7.22 ␦ ␤ CDCl3, ): 34.2 ( carbon of the polymer back- (br, ortho- and para-H from the aromatic ring) ␣ 13 ␦ bone), 40.7 ( carbon of the polymer backbone), ppm. C NMR (CDCl3, ): 27.9–28.0 [(CH3)3C], ␤ 48.5 (OCH3), 174 (carbonyl C from PMA block), 34.9–37.3 ( carbon of the polymer backbone), 177 (carbonyl C from PAA block) ppm. ELEM. 40.3–42.3 (␣ carbon of the polymer backbone),

ANAL. Calcd. for C574H826BrO322 [9300 g/mol, 51.7 (OCH3), 80.4 [(CH3)3C], 126, 128, 145, 174 ϭ ϭ 1 (DPn)PMA 62, (DPn)PtBA 56 by H NMR]: C, (carbonyl C from the PtBA block), 175 (carbonyl C 53.40%; H, 6.40%. Calcd. for C581H824BrO338 from the PMA block). ELEM.ANAL. Calcd. for ϭ ϭ ϭ [16,100 g/mol, (DPn)PMA 90, (DPn)PtBA 80 by C1194H1616BrO236 [20,000 g/mol, (DPn)PMA 62, ϭ ϭ 1 GPC]: C, 52.8%2; H, 6.24%. Found: C, 50.22%; H, (DPn)PtBA 56, (DPn)PS 67 by H NMR]: C, 6.03%. 71.64%; H, 8.08%. Calcd. for C1734H2344BrO340 ϭ ϭ [28,700 g/mol, (DPn)PMA 90, (DPn)PtBA 80, (DP ) ϭ 98 by GPC]: C, 72.50%; H, 8.17%. General Procedure for the Preparation of PtBA-b- n PS Found: C, 72.42%; H, 8.18%. PMA-b-PS in Bulk A 50-mL Schlenk flask fitted with a stir bar (oven- General Procedure for Micelle Formation from dried at 110 °C for 48 h and flame-dried under PAA-b-PMA-b-PS vacuum immediately prior to use) and covered A 500-mL, round-bottom flask fitted with a stir with a rubber septum was charged with PtBA-b- bar was charged with PAA90-b-PMA80-b-PS98 (Mn PtBA-b-PMA-b-PS BY ATRP 4811

ϭ 24,000 g/mol, PDI ϭ 1.16; 120 mg, 0.56 mmol for acrylic acid groups). THF (50 mL, distilled from CaH2) was added, and the solution was al- lowed to stir at room temperature for 10 min to ensure that the mixture became homogeneous. Nanopure water (70 mL, 18 M⍀ cmϪ1) was added via a syringe pump at a rate of 15 mL/h. After all the water was added, the bluish micelle solution Scheme 1 was transferred to dialysis tubing (MWCO ϭ 6000–8000) and dialyzed against deionized wa- ter for 48 h to remove all the THF. The final the preparation of nanostructured materials con- volume was about 240 mL, giving a polymer con- taining hydrophilic and noncrosslinked core do- centration of about 0.51 mg/mL. mains encased within hydrogel-like crosslinked networks. The synthetic design entailed four ba- General Procedure for the Formation of SCK sic steps: (1) the selective cleavage of the tert- Nanostructures from PAA-b-PMA-b-PS butyl ester functionalities afforded an amphiphi- A 250-mL, round-bottom flask fitted with a stir lic block copolymer, (2) the remaining hydropho- bic PMA portion of the chain allowed for the bar was charged with a PAA90-b-PMA80-b-PS98 micelle solution (160 mL, ϳ0.51 mg/mL, 0.30 nucleation of nanoassemblies within an aqueous mmol for acrylic acid groups), followed by 1-[3- solution, (3) the amidation of the acrylic acid res- (dimethylamino)propyl]-3-ethylcarbodiimide me- idues present within the corona of the assemblies thiodide (46 mg, 0.16 mmol, 0.5 equiv to the established a stabilizing intra-assembly crosslinked acrylic acid groups). The mixture was allowed to network, and (4) the hydrolysis of the PMA side stir at room temperature for 10 min to ensure groups converted the core domain into hydro- that the mixture became homogeneous. 2,2Ј-(Eth- philic PAA without detriment to the amide-based ylenedioxy)bis(ethylamine) (23 mg, 0.16 mmol, network linkages. Therefore, PtBA-b-PMA and 0.5 equiv to the acrylic acid groups) was dissolved PMA-b-PtBA were each prepared by sequential in nanopure water (20 mL, 18 M⍀ cmϪ1) and ATRP methods. The addition of PS chain seg- added slowly over 10 min. The solution was al- ments, in the form of the triblock copolymer lowed to stir at room temperature for 12 h before PtBA-b-PMA-b-PS, offered the somewhat surpris- being transferred into dialysis tubing (MWCO ing findings of nonspherical assemblies, and these ϭ 6000–8000) and dialyzed against deionized wa- nanostructures are highlighted in this report. ter to remove all the byproducts. After 6 days, the 2-EBP as an initiator and PMDETA/CuBr as a solution was further dialyzed in nanopure water for catalyst provided reaction conditions under which 1 day. The final volume was about 195 mL, giving a tBA polymerized smoothly at 50 °C in bulk to give polymer concentration of about 0.42 mg/mL. well-defined homopolymers with polydispersities between 1.14 and 1.20 (Scheme 1 and Table I). RESULTS AND DISCUSSION The reaction mixture became progressively more viscous and solidified after about 80 min. Because Block copolymers composed of segments of PtBA the presence of Cu(II) can retard Cu(I)-mediated and PMA were selected as the starting points for ATRP, the control of the amount of Cu(II) is im-

Table I. Homopolymerization of tBA and MA at 50 °C in Bulk

Isolated calc NMR GPC [M]0/ Time Mn Mn Mn Mw/ Conversion Yield a b c Entry Monomer Initiator [I]0 (min) (g/mol) (g/mol) (g/mol) Mn (%) (%) f

1 tBA 2-EBP 100 75 11,000 7,000 12,100 1.14 86 76 0.92 2 tBA 1-EBB 100 85 12,300 9,600 13,500 1.27 96 74 0.91 3 tBA 2-EBiB 100 150 11,000 6,400 12,000 1.22 86 80 0.92 4 tBA 2-ECP 150 20 14,600 — 23,900 1.94 76 — 0.61 5 MA 2-EBP 150 75 8,300 8,700 9,110 1.04 64 55 0.90

a calc ϭ ⅐ ⅐ Mn ([M]/[I]0) (molecular weight of monomer) (% conversion). b Conversion determined from GPC. c ϭ calc GPC f Mn /Mn . 4812 MA AND WOOLEY

Figure 2. The initiators used for the ATRP-based polymerization of tBA included 2-EBP, 2-ECP, 2-EBiB, and 1-BEB.

Figure 1. This composite of GPC eluograms illus- with broader PDIs were obtained. Slower initia- trates the growth of the PtBA chain length and the tion rates were observed with 2-EBiB, requiring concomitant reduction in the relative monomer concen- longer polymerization times than 2-EBP, and the tration after the polymerization of tBA in bulk at 50 °C: resulting polymers had slightly broader PDI val- (a) 55 min, M GPC ϭ 4900, M /M ϭ 1.29; (b) 60 min, n w n ues. In contrast, when 2-ECP was used as an GPC ϭ ϭ GPC Mn 6630, Mw/Mn 1.22; (c) 70 min, Mn ϭ ϭ GPC ϭ initiator together with PMDETA/copper(I) chlo- 8480, Mw/Mn 1.18; and (d) 75 min, Mn 9520, ϭ ϭ ϭ ride, the polymerization of tBA occurred at faster Mw/Mn 1.17. [tBA]0 6.8 M; [2-EBP]0 67.7 mM; [PMDETA] ϭ 67.7 mM; [CuBr] ϭ 67.7 mM. rates. At 50 °C, the reaction mixture became vis- 0 0 cous quickly and solidified within 25 min. GPC analysis revealed a substantial concentration of dead chains. This can be explained by the lower portant. The addition of the monomers to a solubility of CuCl2 causing poor control over the Schlenk flask containing CuBr at the tempera- reversible termination of the active chain ends or ture of liquid nitrogen, followed by degassing the higher bond energy between Cu and Cl shift- prior to the introduction of PMDETA ligand at 77 ing the equilibrium between the dormant and K, minimized the oxidation of Cu(I) to Cu(II) and reactive species toward the reactive species. Ei- produced acceptable polymerization rates. Moni- ther of these situations would lead to an increased toring of the polymerization by GPC showed nice propensity for termination reactions. control over the molecular weight growth and The application of the optimized conditions, polydispersity (Fig. 1). At higher temperatures, involving the use of 2-EBP as an initiator and for example, 90 °C, the polymerization rate of tBA PMDETA/CuBr as a catalyst, to the polymeriza- was too fast; the reaction mixture solidified in less tion of MA also gave smooth polymerizations at than 10 min, and the polydispersity was broad 50 °C in bulk and yielded well-defined PMA ho- ϳ (PDI 2). Therefore, the polymerizations were mopolymers (Scheme 2). GPC analyses showed performed at 50 °C and, after the desired extent of nice control over the molecular weights and poly- conversion was achieved, were rapidly termi- dispersities throughout the polymerization (Fig. nated by immersion of the reaction flask in liquid 3). MALDI-TOF mass spectrometry allowed for nitrogen to give the narrowest molecular weight the determination of the absolute molecular dispersity. The polymer samples were isolated weights and the polydispersities (Fig. 4). The and purified by filtration through a plug of Al2O3 mass difference between the neighboring peaks supported on a bed of celite or SiO2, followed by was 86 Da, corresponding to the repeat unit of repeated precipitations into methanol/water MA. Again, termination of the polymerization via mixtures. SiO2 alone was found to be inefficient the immersion of the flask into liquid nitrogen to for the removal of the copper catalyst from the reaction mixtures, allowing small amounts of copper salts to pass, even after repeated filtra- tions. However, basic Al2O3 on a thin layer of celite or SiO2 effectively removed the copper salts. Three other commonly used initiators, 1-bro- moethylbenzene (1-BEB), 2-EBiB, and ethyl 2-chloropropionate (2-ECP; Fig. 2), were evalu- ated for the initiation of tBA at 50 °C. 1-BEB had the same reaction times as 2-EBP, yet polymers Scheme 2 PtBA-b-PMA-b-PS BY ATRP 4813

Figure 3. This composite of GPC eluograms illus- trates the growth of the PMA chain length and the concomitant reduction in the relative monomer concen- tration after the polymerization of MA at 50 °C in bulk: Figure 5. Polydispersity versus conversion for the GPC ϭ ϭ (a) 20 min, Mn 1580, Mw/Mn 1.13; (b) 45 min, polymerizations of MA (dashed line) and tBA (solid GPC ϭ ϭ GPC Mn 3080, Mw/Mn 1.07; (c) 60 min, Mn line). The insets show the GPC peak shapes for the ϭ ϭ GPC ϭ 5320, Mw/Mn 1.05; and (d) 75 min, Mn 9110, polymers at the highest extents of conversion. ϭ ϭ ϭ Mw/Mn 1.04. [MA]0 10.8 M; [2-EBP]0 72.3 mM; [PMDETA] ϭ 72.3 mM; [CuBr] ϭ 72.3 mM. 0 0 MALDI-TOF agreed with the GPC results ϭ (Mw/Mn 1.04). quench all chain growth at the same time gave The techniques developed for the careful con- the narrowest polydispersity value. The narrow trol of the reaction conditions allowed for the po- ϭ lymerizations of tBA and MA, employing 2-EBP polydispersities (Mw/Mn 1.030) measured by as an initiator and PMDETA/CuBr as a catalyst at 50 °C in bulk, to be performed on relatively large scales (90 g for tBA and 120 g for MA) with results similar to those for the small-scale poly- merizations (5–10 g). During the polymerization of tBA, the polydispersity decreased with increas- ing conversion, and homopolymer coupling was not common. As for MA, the polydispersity was most narrow between 40 and 60% conversions;36 slight increases in the PDI values were observed at higher conversions. In comparison with the polymerizations at higher temperatures, where homopolymer coupling was very common at high conversions, much less homopolymer coupling was observed at 50 °C (Fig. 5). A study of the kinetics of tBA polymerization revealed a linear

increase in ln([M]0/[M]) with time, which is con- sistent with the presence of a constant number of propagating species during the polymerization (Fig. 6). The stereochemistry (tacticity) of the ho- mopolymers was determined by 1H NMR (in

CDCl3) via a comparison of the intensities of the downfield meso proton signal (␦ 1.8–2.0 ppm) with the methine proton resonance (␦ 2.2–2.4 ppm).42 Both PtBA and PMA homopolymers ob- Figure 4. MALDI-TOF spectra acquired from PMA tained under the ATRP reaction conditions were samples dispersed in a matrix of 3-indole acrylic acid atactic (m/r ϭ 50/50). ϭ doped with silver acetate: (a) Mn 6268, Mw/Mn The preparation of PtBA-b-PMA diblock copol- ϭ ϭ ϭ 1.031 and (b) Mn 8598, Mw/Mn 1.030. ymers was accomplished by the polymerization of 4814 MA AND WOOLEY

MA from the PtBA macroinitiators in bulk at 50 °C and in toluene at 90 °C, with no significant difference (Scheme 3). Because toluene is a good solvent for both PtBA and PMA, it was chosen to be the solvent for the solution polymerization, and the conditions employed were similar to those for the homopolymer preparations. As much as 80 vol % toluene could be used to dissolve the PtBA macroinitiator, and although fine particles could be seen toward the late stages of the polymeriza- tion, the heterogeneity did not adversely affect the polymerization. The initiation and propaga- tion also proceeded smoothly in bulk, giving diblock copolymers with narrow molecular weight dispersities (Table II). The bulk polymerizations could be taken to relatively high conversions without polymer chain coupling, unless the poly- merization mixture was allowed to remain at el- evated temperatures for periods of time beyond solidification. MALDI-TOF mass spectra were ob- tained for some of the diblocks (Fig. 7). Because the diblock spectra contained two distributions, having mass differences of 86 and 128 Da due to MA and tBA repeat units, respectively, the peak shapes were not as well resolved as for the PMA homopolymer (Fig. 4). The PDI values calculated from the MALDI-TOF data were similar to those obtained from GPC. A second route for the preparation of diblock copolymers composed of tBA and MA involved the initial polymerization of MA followed by the growth of tBA from the PMA macroinitiator Scheme 3 ϳ (Scheme 4). Because the Tg of PMA is low ( 10 °C), PMA was first frozen in liquid nitrogen and broken into small pieces for convenient measure- was not pumped under vacuum long enough, the ment and transfer. However, if the macroinitiator moisture that condensed during the aforemen- tioned procedure caused a color change of the reaction mixture from the normal green to bright blue. This was easily remedied by the removal of all water from the reaction mixture by azeotro- ping with toluene. The amount of CuBr employed during the polymerization was critical. If an in- sufficient amount was used, in comparison to the molar equivalents of active bromoalkyl chain ends on the macroinitiator, incomplete initiation was observed, and the amount of homopolymer contamination was inversely proportional to the amount of CuBr used. At 1 equiv of CuBr or more, clean initiation was observed, as demonstrated by the GPC eluograms of Figure 8. Selective cleavage of the tert-butyl ester groups Figure 6. Kinetics of the bulk polymerization of tBA in the presence of methyl was achieved by ϭ the diblock copolymers being treated in dichlo- at 50 °C. [tBA]0/[PMDETA]0/[CuBr]0 100/1/1; [tBA]0 ϭ ϭ ϭ 6.8 M; [2-EBP]0 67.7 mM; [CuBr]0 67.7 mM; romethane with anhydrous TFA at room temper- ϭ 1 [PMDETA]0 67.7 mM. ature. The reaction was monitored by H NMR PtBA-b-PMA-b-PS BY ATRP 4815

Table II. Diblock and Triblock Copolymers Prepared by ATRP with Copper(I) Bromide/PMDETA

Entry Macroinitiator [I] Monomer [CuBr]/[I] Solvent T (°C) Time (h) (Mn)expt (Mn)GPC Mw/Mn

1PtBA Styrene 1.1 Toluenea 60 4.5 12,100d 16,100 1.22 2PtBA MA 1.4 Tolueneb 90 2.5 16,800d 18,500 1.11 3PtBA MA 3.0 None 50 2.7 30,217e 32,300 1.19e 4PtBA MA 3.0 None 50 2.0 14,875e 18,500 1.09e 5 PMA t-BA 0.9 Toluenec 90 12.5 — 22,800 1.30 6 PMA t-BA 0.7 Toluenec 90 11.0 — 16,800 1.33 7 PMA t-BA 2.0 Toluenec 90 10.7 12,000d 18,700 1.17 8PtBA-b-PMA Styrene 2.2 Toluenec 90 7.5 — 30,400 1.23 9PtBA-b-PMA Styrene 3.9 None 50 6.7 20,000d 28,700 1.16

a 43% by volume. b 80% by volume. c 50% by volume. d By 1H NMR end-group analysis. e By MALDI-TOF measurement.

(Fig. 9). The signal at ␦ 1.49 ppm, resulting from ppm had nearly disappeared, and the signal from the tert-butyl protons of the PtBA block, de- the PtBA block backbone at ␦ 2.4 ppm shifted creased in intensity, whereas the signal at ␦ 1.62 downfield to ␦ 2.5 ppm. Extended reaction times ppm, corresponding to the tert-butyl trifluoroac- (overnight) were often used to ensure the comple- etate product, appeared and increased in inten- tion of the tert-butyl ester conversion to PAA sity. After 3 h, the tert-butyl ester signal at ␦ 1.49 chain segments. The methyl ester functional groups of the PMA blocks and the end groups of the copolymer remained intact even after 60 h, as was observed in a comparison of their proton res- onance signals and those of the internal standard, bromobenzene. Differential scanning calorimetry was used to evaluate the thermal transition temperatures for

the homopolymers and block copolymers. Tg’s were observed for PMA at 10 °C and for PtBA at 34 °C. The diblock copolymer, PtBA-b-PMA, ex-

hibited two Tg’s at 15 and 44 °C for the immiscible PMA and PtBA phases, respectively. After the conversion of the PtBA block to yield PAA-b-PMA,

Tg’s were observed at 10 °C for PMA and 114 °C for PAA, clearly indicating that the two phases were immiscible. APtBA-b-PMA-b-PS triblock copolymer was also synthesized in bulk at 50 °C in the presence

Figure 7. MALDI-TOF spectra of PtBA-b-PMA diblock copolymers dispersed in a matrix of 3-indole ϭ acrylic acid doped with silver acetate: (a) Mn 14,875, ϭ ϭ ϭ Mw/Mn 1.09 and (b) Mn 30,217, Mw/Mn 1.19. Scheme 4 4816 MA AND WOOLEY

Figure 8. GPC traces illustrate the effects that dif- ferent amounts of CuBr catalyst had on the initiation of 1 tBA as a second block from PMA as a macroinitiator in Figure 9. A time-course H NMR study monitored GPC the selective cleavage of the tert-butyl esters of PtBA- toluene at 90 °C: (a) PMA as a macroinitiator, Mn ϭ ϭ b-PMA via a reaction with TFA (3 equiv with respect to 6610, Mw/Mn 1.04; (b) 0.7 equiv of CuBr to the GPC ϭ the tert-butyl ester groups) at room temperature: (a) macroinitiator to give PMA-b-PtBA, Mn 16,800, ϭ the diblock copolymer in CD2Cl2 (bromobenzene was Mw/Mn 1.33; (c) 0.9 equiv of CuBr to the macroini- GPC ϭ added as an internal standard) at (b) 20 min, (c) 60 min, tiator to give PMA-b-PtBA, Mn 22,800, Mw/Mn ϭ 1.30; and (d) 2 equiv of CuBr to the macroinitiator to (d) 120 min, and (e) 60 h after the addition of TFA. GPC ϭ ϭ give PMA-b-PtBA, Mn 18,700, Mw/Mn 1.17. In addition to spherical particles, cylindrically of CuBr/PMDETA, with the growth of the third shaped nanostructures were formed. The deposi- (PS) block occurring from the diblock copolymer, tion of aqueous solutions containing the mixtures PtBA-b-PMA, bearing a bromoalkyl chain end as of nanospheres and nanocylinders onto a mica a result of the ATRP method of preparation substrate, followed by the sample being allowed (Scheme 5). Under these reaction conditions, sty- to dry freely in air, and an analysis by AFM rene could be polymerized at 50 °C in bulk or at 60 revealed very interesting segregation phenomena °C in toluene, albeit at significantly slower poly- as a function of particle topology. The spheres merization rates than were observed with MA organized into hexagonally packed monolayers in and tBA monomers. Also, substantial amounts of contact with the mica, whereas the cylinders pre- products resulting from polymer chain coupling ferred to assemble on the surface generated by were observed after the polymerization of styrene the spherical particles [Fig. 12(a–c)]. This order- from the PtBA-b-PMA macroinitiator in toluene ing occurred over long ranges, regardless of the at 90 °C when taken to high conversion (Fig. 10). Cleavage of the tert-butyl ester groups of the PtBA-b-PMA-b-PS triblock copolymer was then performed via reaction with anhydrous TFA in dichloromethane at room temperature to afford the amphiphilic PAA-b-PMA-b-PS triblock copol- ymer. A previously reported procedure15 was em- ployed for the assembly of micelles in water, which were then stabilized with a water-soluble carbodiimide, 1-[3-(dimethylamino)propyl]-3-eth- ylcarbodiimide methiodide, to activate the acrylic acid groups present within the shell, followed by crosslinking with a diamine, 2,2Ј-(ethylenedioxy)- bis(ethylamine), to form SCK nanostructured ma- terials (Fig. 11). Scheme 5 PtBA-b-PMA-b-PS BY ATRP 4817

of a variety of shapes and morphologies and also to understand the complex surface interactions that drive their segregation behaviors. Preliminary experiments were performed, uti- lizing the AFM tip as a manipulation tool to ex- amine the mechanical integrity of the SCK nano- cylinders. After tapping-mode AFM imaging was used to visualize the nanostructures, the probe tip was brought into contact with the mica at the position indicated by the arrowhead in Figure 14(a). In this example, a bent cylinder was sub- jected to contact-mode AFM scanning to allow for Figure 10. GPC eluograms of (a) PtBA as the macro- GPC ϭ ϭ the AFM tip to exert lateral force on the nano- initiator, Mn 12,100, Mw/Mn 1.14; (b) PtBA-b- GPC ϭ structure. After the tip manipulation, tapping- PMA as the macroinitiator, Mn 18,500, Mw/Mn ϭ GPC ϭ mode AFM was resumed to collect the image of 1.11; (c) PtBA-b-PMA-b-PS, Mn 30,400, Mw/Mn ϭ 1.23, resulting from the polymerization of the third the same sample region [Fig. 14(b)]. Rather than (PS) segment at 90 °C in toluene; and (d) PtBA-b-PMA- stretching or being moved across the substrate, GPC ϭ ϭ b-PS, Mn 28,700, Mw/Mn 1.16, resulting from the nanostructure instead ruptured. This behav- the polymerization of the third (PS) segment at 50 °C in ior was observed several times and suggests that bulk. the shell thickness and extent of crosslinking in this system did not provide sufficient networking to overcome some combination of the attractive solution concentration of the nanostructures forces with the mica substrate and the local forces [compare the images of Fig. 12(a,b)]. Section anal- applied by the AFM tip. ysis of the AFM images determined that the spheres and cylinders were each of comparable heights (15 Ϯ 2 nm); however, the cylinder CONCLUSIONS lengths extended from 50 nm to 2 ␮m (Fig. 13). Further work is required to control the assembly After intending to produce entirely hydrophilic processes to generate well-defined nanostructures nanostructured materials as stable entities with

Figure 11. The general strategy for the preparation of SCK nanostructures involves the formation of micelles from PAA-b-PMA-b-PS in water and subsequent amidation reactions between the acrylic acid residues of the peripheral chain segments and diamino crosslinkers: (i) 1-[3-(dimethylamino)propyl]-3-ethylcarbodiimide methiodide and 2,2Ј-(ethylenedioxy) bis(ethylamine). 4818 MA AND WOOLEY

complex morphologies within water, we encoun- tered interesting chemistry in (1) the methodol- ogy of ATRP-based techniques for the production of well-defined diblock and triblock copolymers incorporating acrylate monomers, (2) their trans- formation into nanostructured materials of mixed topologies, (3) the topology-driven segregation of those nanostructures, and (4) the evaluation of the mechanical integrity of single nanoscale ma- terials. Further study is required to gain control over the formation of complex nanostructured materials from triblock copolymers, employing

Figure 12. Tapping-mode AFM images of SCK nano- structures that were deposited from an aqueous solu- tion onto mica substrates and then allowed to dry re- veal interesting segregation behaviors. The concentra- tions of the aqueous solutions of the nanostructures were (a) 0.4 mg/mL, (b) 0.08 mg/mL, and (c) 0.08 mg/mL (imaged at 2 ϫ 2 ␮m). Figure 13. Section analysis on an expanded area of the AFM image from Figure 12. (c). A single sphere has a height of 14.9 nm, whereas the combined rod and spheres lying below have a height of 30.4 nm. PtBA-b-PMA-b-PS BY ATRP 4819

8. Sukegawa, M.; Irizato, Y.; Ishitoku, T.; Machida, K.; Fukawa, S.; Katoh, T.; Tamatani, H.; Fukuoka, A. Biodegradable Core–Shell Particles Containing Crosslinked Polyamino Acids and Their Manufac- ture and Use in Hygienic Products or Soil Moistur- izers; PCT Int. Appl. WO 0027904, 2000. 9. Thurmond, K. B., II; Kowalewski, T.; Wooley, K. L. J Am Chem Soc 1996, 118, 7239–7240. 10. Caruso, F.; Schueler, C.; Kurth, D. G. Chem Mater 1999, 11, 3394–3399. 11. Hamers, C.; Rupaner, R.; Leyrer, R. J.; Schuhma- Figure 14. Tapping-mode AFM images of SCK nano- cher, P.; Yang, Z. Pigmented Dispersions of Core– structures on mica, prepared by the deposition of a 0.05 Shell Polymer Particles for Manufacture of Coat- mg/mL solution concentration (a) before manipulation ings; Ger. Offen. DE 98-19834194, 2000. with the AFM tip (the arrow indicates the location at 12. Wooley, K. L. J Polym Sci Part A: Polym Chem which the tip was brought into contact with the mica 2000, 38, 1397–1407. for the mechanical manipulation) and (b) after the 13. Thurmond, K. B., II; Remsen, E. E.; Kowalewski, AFM tip was brought into contact with the mica sub- T.; Wooley, K. L. Nucleic Acids Res 1999, 27, 2966– strate at the position noted by the arrowhead and 2971. scanned to the right in contact mode to exert lateral 14. Huang, H.; Remsen, E. E.; Kowalewski, T.; Wooley, force on the nanostructure. K. L. J Am Chem Soc 1999, 121, 3805–3806. 15. Huang, H.; Wooley, K. L.; Remsen, E. E. Chem the general methodology of self-assembly in solu- Commun 1998, 1415–1416. 16. Wooley, K. L.; Huang, H.; Kowalewski, T. Polym tion followed by covalent stabilization. Advanced Mater Sci Eng 1999, 80, 13. tools will be needed to characterize their surface 17. Zhang, Q.; Remsen, E. E.; Wooley, K. L. J Am interactions and mechanical properties. Addition- Chem Soc 2000, 122, 3642–3651. ally, chemical modification and mechanical ma- 18. Percec, V. Polymerization of Free Radically Poly- nipulation of such nanostructures will be further merizable Vinyl Monomers Using Sulfonyl Halides explored. and Transition Metal Compounds as Initiators; PCT Int. Appl. WO 9820050, 1998. Funding of this research by the National Science Foun- 19. Sawamoto, M.; Kamigaito, M. J Macromol Sci Pure dation (DMR-9974457) is gratefully acknowledged. The Appl Chem 1997, 34, 1803–1814. authors also thank Mr. Yensheng Wang and Professor 20. Matyjaszewski, K.; Patten, T. E.; Xia, J. J Am Michael L. Gross for the MALDI-TOF measurements. Chem Soc 1997, 119, 674–680. The Washington University Mass Spectrometry Facil- 21. Matyjaszewski, K.; Coca, S.; Gaynor, S. G.; Wei, M.; ity is supported by the National Institutes of Health, Woodworth, B. E. Macromolecules 1997, 30, 7348– National Center for Research Resources (Grant No. 7350. P41RR0954). 22. Matyjaszewski, K.; Coca, S.; Gaynor, S. G.; Wei, M.; Woodworth, B. E. Macromolecules 1998, 31, 5967– 5969. REFERENCES AND NOTES 23. Matyjaszewski, K.; Nakagawa, Y.; Jasieczek, C. B. Macromolecules 1998, 31, 1535–1541. 1. Zhang, L.; Eisenberg, A. Science 1995, 268, 1728– 24. Benoit, D.; Chaplinski, V.; Braslau, R.; Hawker, 1731. C. J. J Am Chem Soc 1999, 121, 3904–3920. 2. Zhang, L.; Eisenberg, A. J Am Chem Soc 1996, 118, 25. Hawker, C. J.; Barclay, G. G.; Dao, J. J Am Chem 3168–81. Soc 1996, 118, 11467–11471. 3. Saito, R.; Akiyama, Y.; Tanaka, M.; Ishizu, K. Col- 26. Hawker, C. J.; Barclay, G. G.; Orellana, A.; Dao, J.; loids Surf A 1999, 153, 305–310. Devonport, W. Macromolecules 1996, 29, 5245–5254. 4. Bu¨tu¨ n, V.; Wang, X. S.; Banez, M. V. D. P.; Robin- 27. Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina, J.; son, K. L.; Billingham, N. C.; Armes, S. P.; Tuzar, Jeffery, J.; Le, T. P. T.; Mayadunne, R. T. A.; Z. Macromolecules 2000, 33, 1–3. Meijs, G. F.; Moad, C. L.; Moad, G.; Rizzardo, E.; 5. Bu¨tu¨ n, V.; Lowe, A. B.; Billingham, N. C.; Armes, Thang, S. H. Macromolecules 1998, 31, 5559– S. P. J Am Chem Soc 1999, 121, 4288–4289. 5562. 6. Discher, B. M.; Won, Y.-Y.; Ege, D. S.; Lee, J. C. M.; 28. Chong, Y. K.; Le, T. P. T.; Moad, G.; Rizzardo, E.; Bates, F. S.; Discher, D. E.; Hammer, D. A. Science Thang, S. H. Macromolecules 1999, 32, 2071–2074. 1999, 284, 1143–1146. 29. Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; 7. Sanji, T.; Nakatsuka, Y.; Kitayama, F.; Sakurai, H. Chong, Y. K.; Moad, G.; Thang, S. H. Macromole- Chem Commun 1999, 2201–2202. cules 1999, 32, 6977–6980. 4820 MA AND WOOLEY

30. Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; 36. Wang, J.-S.; Matyjaszewski, K. Macromolecules Krstina, J.; Moad, G.; Postma, A.; Thang, S. H. 1995, 28, 7901–7910. Macromolecules 2000, 33, 243–245. 37. Patten, T. E.; Xia, J.; Abernathy, T.; Matyjasze- 31. Xia, J.; Matyjaszewski, K. Macromolecules 1997, wski, K. Science 1996, 272, 866–868. 30, 7697–7700. 38. Jankova, K.; Kops, J.; Chen, X.; Batsberg, W. Mac- 32. Davis, K. A.; Paik, H.-J.; Matyjaszewski, K. Mac- romol Rapid Commun 1999, 20, 219–223. romolecules 1999, 32, 1767–1776. 39. Davis, K. A.; Matyjaszewski, K. Macromolecules 33. Coca, S.; Jasieczek, C. B.; Beers, K. L.; Matyjasze- 2000, 33, 4039–4047. wski, K. J Polym Sci Part A: Polym Chem 1998, 36, 40. Zhang, X.; Xia, J.; Matyjaszewski, K. Macromole- 1417–1424. cules 2000, 33, 2340–2345. 34. Percec, V.; Barboiu, B.; Neumann, A.; Ronda, 41. Burguiere, C.; Pascual, S.; Coutin, B.; Polton, A.; J. C.; Zhao, M. Macromolecules 1996, 29, 3665– Tardi, M.; Charleux, B.; Matyjaszewski, K.; Vairon, 3668. J.-P. Macromol Symp 2000, 150, 39–44. 35. Wang, J.-L.; Grimaud, T.; Shipp, D. A.; Matyjasze- 42. Liu, W.; Nakano, T.; Okamoto, Y. J Polym Sci Part wski, K. Macromolecules 1998, 31, 1527–1534. A: Polym Chem 2000, 38, 1024–1032.